CpG island methylation and associated silencing is a frequent and early event in colorectal neoplasia. Some of the genes affected, such as MLH1 and p 16, clearly contribute physiologically to the neoplastic phenotype. We now propose the hypothesis that aberrant CpG island methylation will also contribute to the metastatic process in colon carcinoma. By extension, we suggest that identification of CpG islands specifically hypermethylated in metastatic lesions, or primary tumors prone to metastasis, will lead to genes and pathways physiologically involved in the metastasis process. We also propose that the presence of these methylation events in primary tumors will portend a poor prognosis. To test these hypotheses, we propose the following specific aims; (1) To clone CpG islands differentially methylated in metastatic tumors compared to primary tumors, as well as in paired metastatic/non-metastatic primary tumors from patients matched for clinicopathologic characteristics. CpG island cloning will be done using MCA/RDA, a technique developed in this laboratory and verified in multiple tumor types. Current data suggests that 50% of recovered clones are in the promoter of identifiable genes, thus vaIidating this approach for identification of silenced genes and pathways. (2) To screen primary tumors for aberrant methylation of the most promising clones, and relate methylation to prognosis in a group of tumors well characterized genetically and epigenetically. (3) To begin exploring the function of recovered genes using in-vitro transfection experiments. Successful completion of this project will lead to a better understanding of the metastatic process in colon cancer, better evaluation of the metastatic potential of primary tumors, new markers of prognosis in colon cancer and will lay the groundwork for the use of methylation inhibitors in the prevention and treatment of metastasis in colorectal neoplasia.